Did homicides increase after the UK banned handguns?

Now, hold your horses before getting fired up on this topic. I’m not here to debate gun control, I’m only here to debunk a single myth that I hear time and time again about my country. It goes as follows:

After the United Kingdom banned handguns and tightened gun restrictions in 1997, homicides actually increased.

This is usually paired with the following chart from the British government:

Now there is no false data recorded in this graph, and at first glance you see a homicide spike followed by a gradual reduction in homicides after the ban. The previous trend was a steady increase in homicides, and now it does seem that we are having a steady decrease in homicides.

But what’s with the spike? Did gun control actually cause homicides to increase by 50% for a decade?

There is a missing detail

All homicides recorded are plotted on this graph, but the UK policy for recording homicides after 2003 is to plot the homicides on the year they were discovered, not when they were committed.

The spike you see is the work of none other than Dr. Harold Shipman, possibly the world’s most prolific serial killer, who performed his killings in 1985. The UK homicide rate is so low (500-100 persons a year) that Harold Shipman appears as a large spike as he had 172 confirmed victims plotted here and was estimated to have killed up to 250 people. He killed so many people that Wikipedia created a section for medical professional serial killers to distinguish the kills from normal murders, because he was able to kill in such great numbers.

Some other peaks and spikes on the chart include some of the following events that either happened before 1997 or are anomalous events:

Perhaps Derrick Bird is the most debatable as to whether he should be on the list. Terrorist attacks are also anomalous, it would be like talking about 9/11 when debating school shootings.

The graph with killings put back into their relevant times and anomalous events adjusted looks like this:

There is an increase in the years 2001-2005, but you see that the spiking upward trend is no longer present with shipman’s killings in the 1980s taken out. But look closely and you will see that there is no great increase in the years 1997, 1998, 1999 or 2000, which is when we would expect to see an effect. To make this point clearer, if you look at the period of 1977-1979 that has a greater increase in homicides than 1998-2000, so we ultimately do not see any effect at all as a result of the ban.

Now, there is a downward trend of homicides and we will have to wait another 20 years to see how it gets on, but this blog post is only about a single thing: premature conclusions. This topic is extremely complicated and you would do well to analyse the statistics in detail and also understand that it’s very hard to nail down causes on such a broad statistic being measured.


I would like to thank the users of the Skeptics Stack Exchange for their response to this question, where they created the adjusted graph and collected all the information for easy viewing, particularly the great effort put in by the user Schwern. I highly recommend this website for fact checking.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.